Information Feedback Based Architecture for Handling the Scalability Issues in the Reusable Cloud Components
With the evolution of the paradigm of cloud computing in every field of application, the demand on the reusable resources while providing the service has increased substantially. Although it is transparent to the user through virtualization, that is also a strength of cloud computing, the runtime scalability of resources to cater for a variety of services is to be addressed to meet the critical factor of the agreed quality of service. In this work, an architecture based on information feedback is presented to address this issue. The findings have been supported by the simulation results. The scalable architecture makes use of a hierarchy of resources, each level capable of providing a different degree of services. The demand for resources at each level, which is also equivalent to contention for resources or service drops, is computed using Random Early Detection (RED) or similar algorithms and used as feedback signal. The effectiveness of this signal may be enhanced by predicting the same several steps ahead of time. The prediction ensures the availability of a breathing time for the allocation of the resources.